| Book Review | The Journal of American History, 90.2

You have not been recognized as a subscriber to JAH online. About 188 words from this article are provided below; about 357 words remain.

If you are a individual member of the Organization of American Historians, you may:: • login here if you have already registered for online access.; • Or if you're already logged in register your subscription.
• Set up your online account for the first time.
If you are not a member of the Organization of American Historians, you can:: • Join the OAH and receive many member benefits including print and electronic issues of the Journal of American History.
• Purchase a research pass to gain two-hour access to the entire History Cooperative web site. You will have full access to current issues of the Journal of American History (86.1-present). Note: the Research Pass does not provide access to JSTOR's holdings of the Journal of American History.
Instititutions can:: • Subscribe to this journal and receive print and electronic issues.
• Activate your existing subscription so that we recognize your IP number ranges.

| Book Review | The Journal of American History, 90.2 | The History Cooperative

Book Review

Who Counts? The Politics of Census-Taking in Contemporary America. By Margo J. Anderson and Stephen E. Fienberg. (New York: Russell Sage Foundation, 1999. x, 319 pp. Cloth, $32.50, ISBN 0-87154-256-0. Paper, $16.95, ISBN 0-87154-257-9.)

Who Counts? evaluates the efforts of the Bureau of the Census to improve and correct its population enumeration, particularly after 1950, and the methodological debates and the politicization of the dispute since 1980. The book begins with a brief history of the census, drawn from the coauthor Margo J. Anderson's monograph, The American Census (1988). The most important points are that the census never took an actual head count of every person, that sampling has been used since 1940, and that the political implications of census results increased after the mid-1960s because of struggles over reapportioning legislatures, allotting federal funds, and evaluating discrimination. After discovering in the early 1940s that the census undercounted the population, the bureau struggled to improve its coverage. Accuracy steadily improved, yet the 1980 and 1990 censuses still included roughly 1–2 percent net error, 7 percent total error, and 6 percent undercount of the black population.

. . .

There are about 357 more words in this article. Please log in (or, if you are not yet an authorized user, please go to the User Setup page) to gain full access rights. Or if you're already logged in register your subscription.